Essentially all CNS lymphomas, as well as most peripheral lymphomas, in AIDS carry the EBV genome. We hypothesize that novel approaches that specifically target EBV-infected cells for destruction will be useful for the treatment of AIDS-related lymphomas. In this grant, we propose to develop and compare different EBV-based approaches for the treatment of AIDS-related lymphomas, capitalizing upon our extensive knowledge of EBV gene regulation. In our first specific aim, we will investigate the feasibility of using lytic EBV infection, in combination with cytotoxic prodrugs that are activated by lytic EBV gene products, to kill EBV-associated lymphomas. We will identify the most efficient agents for inducing lytic EBV infection in lymphomas, investigate the underlying viral and cellular mechanisms responsible for this effect, and compare the efficacy of ganciclovir versus AZT (in combination with lytic inducing agents) for treating EBV-positive lymphomas in vivo. In our second specific aim, we will determine if the cellular CD70 protein (the ligand for the CD27 receptor), can be used to direct EBV-specific killing of cells. CD70 is over-expressed on the surface of most EBV-positive malignancies, but very few normal cells. We will also examine the importance of the CD70/CD27 interaction for regulation of EBV gene expression, as well as cell growth, in EBV-positive lymphoblastoid cells in vitro and in vivo. In our third specific aim, we will investigate the role of the EBV immediate-early proteins, if any, in EBV-induced lymphomas in vitro, as well as in vivo, using BZLF1 and BRLF1 knock-out viruses. Our preliminary data suggest that BZLF1 expression may be required for efficient outgrowth of EBV-infected primary B cells in vitro, possibly due to a BZLF1 -induced paracrine factor. Our proposed studies may lead to novel, EBV-based strategies for treating AIDS-related lymphomas. [unreadable] [unreadable] [unreadable]